Mitigating Information Manipulation of AI Influencers
Artificial Intelligence (AI) influencers’ wider adoption makes addressing the risks of perceived information manipulation crucial. This study employs a mixed-methods approach to investigate the antecedents and consequences of perceived information manipulation by AI influencers. Drawing on Hunt-Vitell Theory and Goal-Framing Theory, a theoretical model is tested using survey data from 625 participants in Singapore. The results indicate that reasonable use, privacy protection, non-deception, personal development, and perceived enjoyment significantly reduce perceived information manipulation, while informed consent has a negative but non-significant effect. Perceived information manipulation negatively impacts both routine and infusion use. Thematic analysis of 15 follow-up semi-structured interviews generates meta-inferences that provide both confirmation and complementary insights into the psychological mechanisms underlying the quantitative results. Based on these findings, this study offers practical recommendations for AI influencer operators and policymakers to mitigate potential negative social impacts, enabling the ethical realization of AI influencers’ full potential.
Authors
- Kum Fai Yuen (ORCID: https://orcid.org/0000-0002-9199-6661)
- Zhaotong Li (ORCID: https://orcid.org/0009-0001-7234-3446)
- Chee-Chong Teo
Institutions
- Nanyang Technological University (SG)
Publication Details
- Journal
- International Journal of Human-Computer Interaction
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1080/10447318.2026.2760674
- Primary Topic
- Digital Marketing and Social Media
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Ministry of Education - Singapore